JP5135205B2 - Acoustic compression encoding apparatus and decoding apparatus for multi-channel acoustic signals - Google Patents

Description

  The present invention relates to an acoustic compression encoding apparatus and decoding apparatus for multi-channel acoustic signals, and in particular, in an acoustic system composed of a large number of channels such as a three-dimensional (stereoscopic) acoustic system, the positional relationship between the channels and the acoustic signals. The present invention relates to an acoustic compression encoding apparatus and decoding apparatus for a multi-channel audio signal that can reduce the transmission bit rate and the encoding load as compared with the related art by encoding using the similarity (correlation) between them.

  Currently, there are many known low-bit-rate acoustic compression coding technologies that can transmit high-quality sound signals that are inferior to the original sound at a transmission bit rate significantly lower than that of the original sound (Linear PCM: Pulse Code Modulation). ing. For example, there is one that encodes at a bit rate of about 1/10 or less of the original sound (linear PCM) using an auditory property such as masking (see Non-Patent Document 1, for example). Even if such encoding is performed, the difference between the encoded sound and the original sound is hardly perceived by hearing. In addition, as encoding of 5.1 channel surround sound or the like, stereo which is encoded at a bit rate smaller than “required bit rate of single channel signal × number of channels” by using the similarity of two left and right channel acoustic signals. There is a compression encoding method (see, for example, Patent Document 1). In addition, there is a lossless encoding method that enables the original sound on the receiving side to be restored and the transmission bit rate to be reduced by predicting the acoustic signal of a certain channel on the transmitting side from the acoustic signal of another channel and sending a prediction error signal. (For example, see Non-Patent Document 2 and Non-Patent Document 3).

  On the other hand, a “three-dimensional audio system” suitable for large-screen and high-definition video systems capable of reproducing sound with a higher presence than the 5.1 channel surround sound system has been developed (for example, non-patent literature). 4). This expands the channel hierarchically in the height direction in addition to the plane, and enables sound image localization in the height direction. This three-dimensional sound method will be described with reference to the drawings. FIG. 8 is an example of speaker (channel) arrangement at the time of program production in the three-dimensional sound system. As shown in FIG. 8, in the production of a three-dimensional audio system program, the program is produced under standard production conditions in which a large screen video display (for example, 7680 × 4320 pixels) and a large number of speakers are arranged. That is, under the standard production conditions, the large-screen display DIS is moved forward, except for the low-frequency sound effect speakers LFE1 and LFE2, the three-dimensional upper layers U1 to U9, the intermediate layers M1 to M10, and the lower layers L1 to L3. The sound signal is produced in 22 channels.

JP 2002-268694 A ISO / IEC13818-7, 2003 ISO / IEC14496-3, 2005 / AMD2 Kamamoto, Moriya, Nishimoto, Kashiyama, “Lossless compression coding of multi-channel signals using inter-channel correlation”, IPSJ Journal, vol. 46, No. 5, pp. 1118-1128, May 2005 Akio Ando, “High Realistic Acoustics”, NHK Broadcasting Technology Laboratory, [Search Date: December 15, 2008], Internet <http://www.nhk.or.jp/strl/group/ningen_joho/ningen_joho06 .html>

  This three-dimensional sound system is composed of a large number of sound signals (in the case of FIG. 8, the main channel is 22 channels and the low-frequency emphasis channel is 2 channels). Therefore, a large bit rate is required to transmit these acoustic signals as the original sound. In addition, many acoustic signals of this three-dimensional acoustic system have stronger similarity between acoustic signals, in particular, between acoustic signals of adjacent acoustic channels, compared to two-channel stereo sound and 5.1-channel surround sound system. However, a low bit rate that utilizes the strong similarity of many sound channel signals of this three-dimensional sound system, that is, takes advantage of the characteristics of the channel arrangement in which a plurality of channels are arranged adjacent to each other, such as the three-dimensional sound system. An acoustic signal compression coding technique has not been developed.

  Accordingly, an object of the present invention is to provide a multi-channel sound system that uses a large number of sound channels, such as a three-dimensional sound system that can reproduce sound with a higher presence than the 5.1-channel surround sound system. A technique (encoding device, decoding device) relating to low bit rate acoustic compression encoding that can be transmitted at a transmission bit rate smaller than “required bit rate of single channel signal × number of channels” while maintaining high acoustic quality. ) To provide.

In order to solve the above-described problems, an acoustic compression encoding apparatus for a multi-channel acoustic signal by a plurality of acoustic channels arranged apart (three-dimensionally or two-dimensionally) according to the present invention is provided.
A grouping unit that groups the plurality of acoustic channels based on predetermined acoustic channel arrangement information (information indicating the positional relationship of the channels);
For each group grouped by the grouping unit, a category dividing unit that divides the acoustic channels belonging to the group into categories based on the correlation of the acoustic signals of the acoustic channels;
For each category divided by the category dividing unit, each acoustic signal belonging to the category is encoded, and an acoustic compression coded signal for each category obtained by multiplexing the encoded acoustic signals is output. An encoding unit;
And a multiplexing unit that multiplexes and transmits the category-specific acoustic compression encoded signals output by the category acoustic compression encoding unit.

An acoustic compression encoding apparatus for a multi-channel acoustic signal according to an embodiment of the present invention is provided.
The received data is included in a single frame;
The category dividing unit divides the acoustic signals of the acoustic channels belonging to the group into categories for each unit frame for encoding the acoustic signals.

An acoustic compression encoding apparatus for multi-channel acoustic signals according to another embodiment of the present invention includes:
It further comprises a bit rate distribution unit that distributes a predetermined bit rate to each category divided by the category division unit.

An acoustic compression encoding apparatus for a multi-channel acoustic signal according to still another embodiment of the present invention includes:
The bit rate distribution unit distributes a predetermined bit rate according to the number of channels belonging to each category.

As described above, the present invention has been described as the encoding device on the transmission side. However, the present invention can also be realized as a decoding device that decodes an acoustic signal transmitted by these encoding devices. It should be understood that these are included in the scope.
For example, an acoustic decoding device that implements the present invention as a decoding device and receives and decodes multi-channel acoustic signals from a plurality of acoustic channels arranged apart (three-dimensionally or two-dimensionally) from each other,
The multi-channel acoustic signal is a multiplexed signal in which acoustic compression-coded signals for each category encoded for each category based on the correlation of the acoustic signal are multiplexed.
A separation unit that separates the multiplexed signal into acoustic compression-coded signals for each category, and extracts category division information related to the separated categories;
A category-specific acoustic decoding unit that decodes the category-specific acoustic compression encoded signals separated by the separation unit for each category and outputs each acoustic signal belonging to the category;
Wherein based on the category division information separated by the separating unit integrates the acoustic signal output by the category sound decoding unit, characterized in that it comprises a category integration unit for output force.

  The apparatus according to the present invention can also be configured as a computer having a central processing unit (CPU, etc.), a memory, etc., and the above-described units are realized by causing the CPU to execute a program describing the function of each unit. can do.

  As described above, according to the present invention, in a multi-channel sound method using a plurality of sound channels such as a three-dimensional sound method, the encoding load is reduced as compared with the conventional method and “required bit rate of a single channel signal × channel” Transmission at a transmission bit rate smaller than "number" is possible.

  Hereinafter, an acoustic compression encoding apparatus and decoding apparatus for a multi-channel acoustic signal according to an embodiment of the present invention will be described with reference to the drawings. In the following description, the acoustic channel arrangement shown in FIG. 8 is assumed. However, the present invention is not limited to this, and the present invention can also be applied to a case where the number of arranged channels is smaller (for example, two channels) or a large number of channels are arranged on a plane. Please note that. Further, hereinafter, “acoustic channel signal” refers to an acoustic signal generated from each acoustic channel, and may be used synonymously with “acoustic signal”. Further, “multi-channel acoustic signal” is used as a general term for a plurality of “acoustic channel signals”.

  First, FIG. 1 shows a schematic diagram of an acoustic signal transmission system including an acoustic compression encoding apparatus and decoding apparatus for multi-channel acoustic signals according to the present invention. As shown in FIG. 1, the acoustic signal transmission system includes a multi-channel sound production / operation device 300 and multi-channel sound compression / encoding device 100 such as three-dimensional sound as a transmission side (broadcast station), and a multi-channel sound decoding device as a reception side. 200 and a multi-channel sound reproduction apparatus 400 such as a three-dimensional sound. The multi-channel sound production / operation device 300 such as a three-dimensional sound is a multi-channel sound signal included in a program content produced in a format in which channels are three-dimensionally arranged in a three-dimensional space as shown in FIG. Based on the program operation of the broadcasting station, the data is sent to the multi-channel audio compression coding apparatus 100. The multi-channel audio compression encoding apparatus 100 encodes the multi-channel audio signal transmitted from the multi-channel sound production / operation apparatus 300 such as three-dimensional sound, and generates an audio encoded stream signal with a low bit rate (details will be described later). To do). The stream signal generated by the multi-channel acoustic compression encoding apparatus 100 is multiplexed with the image data stream signal of the program content by a multiplexing apparatus (not shown) and transmitted to the receiving side wirelessly or by wire.

  On the receiving side (general household, company, etc.), the multi-channel audio decoding device 200 decodes the audio encoded stream signal separated from the image data stream signal by a separation device (not shown) into a multi-channel audio signal format. The multi-channel sound signal decoded by the multi-channel sound decoding device 200 is sent to the multi-channel sound reproduction device 400 such as a three-dimensional sound and reproduced.

  Next, a multi-channel acoustic compression coding apparatus according to the present invention will be described. FIG. 2 is a schematic block diagram of a multi-channel acoustic compression encoding apparatus 100 according to an embodiment of the present invention. As shown in FIG. 2, the multi-channel audio compression encoding apparatus 100 includes a multi-channel audio category division unit 110, an audio compression encoding unit 120, a multi-channel audio multi-category encoded stream multiplexing unit 130, and each category allocation bit rate. A distribution unit 140 is provided. Furthermore, the acoustic compression encoding unit 120 includes the number of categories n (n is the number of categories divided by the multi-channel acoustic category division unit 110 such as the intra-category 1 acoustic channel compression encoding unit 121 and the intra-category 2 acoustic channel compression encoding unit 122). Category-n acoustic channel compression encoding unit 12n corresponding to (natural number excluding 0).

  The multi-channel sound category division unit 110 receives the multi-channel sound signal sent from the multi-channel sound production / operation apparatus 300, and groups the sound channels into a plurality of groups based on the channel arrangement. Furthermore, the multi-channel acoustic category division unit 110 divides (clusters) the acoustic channel signals included in the group into a plurality of categories based on the similarity of the acoustic channel signals for each group. Also, information related to category division (multi-channel audio category division information) is output to the multi-channel audio multi-category encoded stream multiplexing unit 130 and each category assignment bit rate distribution unit 140.

  The acoustic compression encoding unit 120 encodes an acoustic channel signal belonging to a category for each category divided by the multi-channel acoustic category division unit 110, and multiplexes the encoded acoustic channel signals for each category. An encoded signal is generated. That is, the acoustic compression encoding unit 120 generates the same number of category-specific acoustic compression encoded signals as the number of categories divided by the multi-channel acoustic category division unit 110.

  The multi-channel acoustic multi-category coded stream multiplexing unit 130 includes a plurality of category-specific acoustic compression encoded signals generated by the acoustic compression coding unit 120 and the multi-channel acoustic category dividing unit 110. Information is multiplexed into a single transmission stream. As will be described in detail later, multi-channel acoustic category division information includes information on category division of multi-channel acoustic signals, encoding information of acoustic channels in categories, and information on multiplexing of acoustic channel encoding information in categories. Etc. are included.

  Each category assignment bit rate distribution unit 140 assigns a predetermined required bit rate to each category based on the multi-channel sound category division information from the multi-channel sound category division unit 110.

  Note that the encoding of the acoustic signal by the multi-channel acoustic compression encoding apparatus 100 is performed for a predetermined unit frame or for several consecutive frames. This frame is configured to be able to synchronize with the video signal. For example, in the case of synchronizing between the audio signal 48 kHz sampling and the video signal 30 Hz, the frame is composed of 1600 audio samples.

  Detailed functions of each part will be described below. FIG. 3 is a schematic block diagram of the multi-channel acoustic category division unit 110. As shown in the figure, the multi-channel acoustic category division unit 110 includes a multi-channel grouping unit 112 and a category division unit 114. The category division by the multi-channel acoustic category division unit 110 is based on the following concept.

  First, when the conventional coding using auditory properties such as masking is applied to the three-dimensional sound method, the channel is installed in the front, rear, and vertical directions in the three-dimensional sound method. The degree of making it difficult to hear (masking) the sound in the same direction is different from the degree of making it difficult to hear the sound in another direction. That is, it is necessary to calculate the masking amount according to the channel positional relationship. In other words, it is considered that a plurality of channels located at the same position with respect to a certain channel have the same degree of masking from the certain channel. Therefore, if the channels are grouped in advance based on the positional relationship of the channels and the same masking amount is applied to the grouped channels, the process of calculating the masking amounts for a plurality of channels can be reduced.

  In addition, when the transmission bit rate is reduced by encoding based on the similarity between acoustic signals according to the prior art, the similarity of the acoustic signals is largely due to the positional relationship of the acoustic channels. It is considered that the similarity is greater between the acoustic channel signals in the group based on the. That is, it is considered that the advantage of the encoding based on the similarity between all the acoustic channel signals can be sufficiently enjoyed only by performing the encoding based on the similarity between the acoustic channel signals in the group. Therefore, the grouping process can reduce the similarity inspection process rather than the similarity inspection for all the acoustic channel signals.

  Returning to FIG. 3, the function of each unit will be described. The multi-channel grouping unit 112 groups multi-channel acoustic signals based on arrangement information of a large number of acoustic channels. The grouping method can be performed according to the relative position (front, back, side, etc. of the viewer) of the channel with respect to the most ideal viewer position (for example, several meters from the screen). Further, an absolute distance from the viewer may be included. The multi-channel arrangement information is input in advance by the program producer. Alternatively, a predetermined test signal may be transmitted and received between acoustic channels, and the position between channels may be calculated based on the reception time of the test signal.

  FIG. 4 shows an example of grouping. For the sake of simplicity, two-dimensional channel grouping will be described. As shown in FIG. 4, ten channels M1 to M10 are arranged for the viewer AU. In the example of FIG. 4, the five channels M1 to M5 located in front of the viewer AU are one group G1, the three channels M6 to M8 located behind are one group G2, and M9 and M10 located on the left and right. Are grouped into groups G3 and G4, respectively. It should be noted that the figure is an example, and the present invention is not limited to this. For example, as the grouping, a penalty function method related to constrained optimization can be used.

  For each group grouped by the multi-channel grouping unit 112, the category dividing unit 114 divides the channels included in the group into at least one category based on the similarity of the acoustic channel signals. As a division method of this category, an existing method can be used, for example, hierarchical clustering, prediction difference minimizing clustering, or the like can be used (for example, Toriwaki, “Cognitive Engineering”, Corona, pp. 84). -95, Nagao, "Pattern Information Processing", Corona, pp. 98-124, see Non-Patent Document 3). Other known Max-Min methods can also be used. In the example of FIG. 3, the signal is divided into n categories. At this time, signals of different groups do not belong to the same category. The category dividing unit 114 sends out the acoustic channel signals belonging to each divided category to the subsequent acoustic compression encoding unit 120. For example, a flag corresponding to the number of categories is prepared, and the flag is used to indicate an acoustic channel signal belonging to category n (category n acoustic channel signal). The category dividing unit 114 also generates information related to category division (the number of categories, the number of acoustic channels included in each category, and the like), and sends the generated information to the subsequent acoustic compression encoding unit 120.

  The acoustic compression encoding unit 120 encodes each category n acoustic channel signal for each category. Prior to the acoustic compression encoding unit 120, each category allocation bit rate distribution unit 140 will be described.

  Each category assignment bit rate distribution unit 140 distributes a predetermined (required) bit rate to the categories divided by the multi-channel acoustic category division unit 110. Since the acoustic signal and its masking threshold vary from moment to moment, if the bit rate is adaptively allocated according to the variation, the acoustic signal can be further improved in sound quality. However, in this case, the amount of calculation processing and time required for compression encoding are enormous. Therefore, in the present application, a method of allocating the bit rate according to the number of acoustic channels included in the category is adopted in consideration of a necessary calculation processing amount and sound quality.

  As an example, it is conceivable to allocate the bit rate according to the number of acoustic channels included in the category. That is, for each category, the same bit allocation is performed on the first acoustic signal, and the second and subsequent acoustic signals are allocated a bit number obtained by exponentially subtracting the first allocated bit number. At this time, the ratio of the number of allocated bits allocated to the jth category is given by the following equation, where ch is the number of acoustic channels included in this category.

  Furthermore, if the overall bit rate is br_total and the number of category divisions is N, the number of allocated bits assigned to the jth category is given by the following equation.

  Note that the value of α may be, for example, α = ½.

  In this way, based on the property that the similarity between the acoustic signals in each category is high, while the similarity between the acoustic signals in different categories is weak (no), it is required according to the number of acoustic channels in the category. Can be allocated. Each category allocation bit rate distribution unit 140 outputs the bit rate distribution obtained as described above to the acoustic compression encoding unit 120.

The function of the acoustic compression encoding unit 120 will be described. The acoustic compression encoding unit 120 includes a category-n acoustic channel compression encoding unit 12n that encodes the category-specific acoustic channel signals divided by the multi-channel acoustic category division unit 110, respectively. In addition, in order to make it easy to understand conceptually, in FIG. 2 , the n intra-category acoustic channel compression encoding units are shown. Note that it is not provided individually. Therefore, the configuration of the category 1 acoustic channel compression encoding unit 121 to the category n acoustic channel 12n is the same.

  FIG. 5 shows a schematic block diagram of the intra-category 1 acoustic channel compression encoding unit 121. As shown in the figure, the category 1 acoustic channel compression encoding unit 121 includes a time-frequency conversion unit 121A, a requantization / coding unit 121B, a masking threshold calculation unit 121C, a category 1 threshold calculation unit 121D, and an acoustic channel code. A multiplexed information multiplexing unit 121E and a category allocation bit rate storage unit BIT1. The time-frequency conversion unit 121A performs frequency conversion on the time-domain acoustic channel signal included in the category 1. The requantization / encoding unit 121B requantizes and encodes the acoustic channel signal converted into the frequency domain by the time-frequency conversion unit 121A. The masking threshold value calculation unit 121C calculates a masking threshold value for each acoustic channel signal included in the category 1, and the category 1 threshold value calculation unit 121D sets the masking threshold value for each acoustic channel obtained by the masking threshold value calculation unit 121C. Based on this, a masking threshold for the entire category 1 is calculated.

  The masking threshold value calculation unit 121C calculates a masking threshold value (individual channel masking threshold value) for each acoustic channel signal along the auditory critical bandwidth or frequency subband for each acoustic channel signal included in the category. The category 1 threshold calculation unit 121D calculates a masking threshold (category masking threshold) for a category using the individual channel masking threshold obtained by the masking threshold calculation unit 121C. For example, the category 1 threshold value calculation unit 121D may obtain the maximum one of the individual channel masking threshold values obtained by the masking threshold value calculation unit 121C as the category masking threshold value. Alternatively, the category masking threshold can be obtained from the sum of the power of each individual channel masking threshold. The calculation method of the individual channel masking threshold is known from, for example, ISO / IEC 11172-3, and the description thereof is omitted.

  As a method of converting a time domain acoustic signal by the time-frequency conversion unit 121A into a frequency domain signal (frequency sample signal), a method of dividing into a plurality of bands by subbands and DCT (Discrete Cosine Transform) directly. There is a method of converting to frequency. MPEG-1 Layer 2 encoding is used as the subband method, and MPEG-2 AAC (see ISO / IEC 13818-7) is used as the DCT conversion method. For these time-frequency conversions, for example, ISO / IEC 11172-3 is used. The description is omitted because it is already known.

  The requantization / encoding unit 121B adaptively allocates the number of bits allocated in advance by each category allocation bit rate allocation unit 140 to the acoustic channel signal in the category for each critical band (or frequency subband). . For example, the bits may be assigned in the order of the critical band in which the category masking threshold obtained by the category 1 threshold calculation unit 121D and the difference in the amplitude level (decibel difference) of the acoustic channel signal are large. In this case, bit allocation ends when the number of bits allocated by each category allocation bit rate distribution unit 140 is exceeded. Furthermore, the requantization / encoding unit 121B requantizes and encodes the frequency sample values included in the frequency subband based on the number of bits allocated to the critical band. Here, in order to further reduce the number of transmission bits of the encoded frequency sample value, lossless encoding such as Huffman code can be used. In addition, since requantization and encoding are known from, for example, ISO / IEC 11172-3, detailed description is omitted.

  The acoustic channel encoded information multiplexing unit 121E multiplexes the acoustic channel signals in the category encoded by the requantization / encoding unit 121B and the encoded information related to the acoustic channel signals into one signal stream. . The encoded information includes critical band bit allocation information of the acoustic channel signal, frequency sample encoded data included in a frequency subband encoded based on the critical band bit allocation information, and the like.

  Moreover, when performing the acoustic compression encoding in a category, prediction encoding can also be used (for example, refer nonpatent literature 3). Category division is performed based on the similarity of the acoustic channel signals. Therefore, by re-quantizing and encoding the prediction residual of the acoustic channel signal within the category, the number of transmission bits is smaller than when re-quantizing and encoding the acoustic channel signal as it is (conventional encoding). Thus, sound quality equivalent to that obtained when the acoustic channel signal is re-quantized and encoded as it is can be achieved. In other words, if the number of transmission bits is the same, the use of predictive coding can provide higher sound quality than conventional coding. However, calculation of the prediction residual requires decoding / difference / requantization processing, which increases the amount of calculation processing. Therefore, it is only necessary to determine whether or not to use predictive coding in consideration of the necessary number of bits of the auxiliary signal for prediction (additional information) and the increase in the processing amount.

  In this case, as encoded information multiplexed by the acoustic channel encoded information multiplexing unit 121E, in addition to the above, information on the presence / absence of predictive encoding, information on predictive encoding, and frequency sample code for predictive encoding It is also possible to include data.

  As described above, according to the acoustic compression coding apparatus according to the present invention, a multi-channel acoustic signal using a large number of acoustic channels such as a three-dimensional acoustic system is based on the positional relationship between channels and the similarity between acoustic signals. Therefore, the transmission bit rate and the load of the encoding process can be reduced as compared with the conventional case. In addition, since bits are allocated according to the number of channels included in the category, it is possible to transmit without degrading the quality of the acoustic signal.

  Next, an audio decoding device that decodes an audio compression encoded stream signal generated by the above audio compression encoding device will be described. In other words, the acoustic decoding device according to the present invention is a device that decodes a signal in which a plurality of acoustic channel signals are encoded for each category based on their correlation and a multiplexed acoustic compression signal for each category is multiplexed. .

  FIG. 6 is a schematic block diagram of a multi-channel audio decoding device according to an embodiment of the present invention. As shown in the figure, the multi-channel acoustic decoding device 200 includes a multi-channel acoustic multi-category encoded stream separation unit 210, an acoustic decoding unit 220, and a multi-channel acoustic category integration unit 230. The multi-channel audio multi-category encoded stream separation unit 210 separates the encoded stream signal transmitted by the multi-channel audio compression encoding apparatus 100 into the audio channel encoding information and the audio channel signal encoded for each category. To do. Furthermore, the separation unit 210 also separates information related to category division (multi-channel acoustic category division information). The acoustic decoding unit 220 includes an acoustic channel acoustic decoding unit 22n in the category n, decodes the acoustic compression encoded signal for each category separated by the separation unit 210 for each category based on the encoding information, and is included in the category Each acoustic channel signal is decoded. The multi-channel acoustic category integration unit 230 integrates the acoustic channel signals included in each category decoded by the acoustic decoding unit 220 based on the multi-channel acoustic category division information, and restores the multi-channel acoustic signal.

  FIG. 7 shows a schematic block diagram of the category 1 acoustic channel acoustic decoding unit 221. As illustrated, the intra-category 1 acoustic channel acoustic decoding unit 221 includes an intra-category 1 acoustic channel encoded stream separation unit 221A, a decoding unit 221B, and a frequency-time conversion unit 221C. The category 1 acoustic channel encoded stream separation unit 221A converts the acoustic compression encoded signal related to category 1 separated by the separation unit 210 into encoded signals of individual acoustic channel signals included in category 1 (acoustic channel encoded signal). ). Each acoustic channel encoded signal is output as a time-domain acoustic channel signal by the decoding unit 221B and the frequency-time conversion unit 221C. Note that the processing of the acoustic decoding device 200 and each component thereof has the reverse configuration of the acoustic compression encoding device 100, and detailed description thereof is omitted.

  According to the present invention, in a multi-channel acoustic system using a plurality of acoustic channels such as a three-dimensional acoustic system, encoding processing is performed more than before by encoding using a positional relationship between channels and similarity between acoustic signals. Therefore, it is possible to transmit at a transmission bit rate smaller than “required bit rate of a single channel signal × number of channels”, which is useful for multi-channel audio systems.

  Although the present invention has been described based on the drawings and examples, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these variations and modifications are included in the scope of the present invention. For example, functions and the like included in each component can be rearranged so as not to be logically contradictory, and a plurality of components can be combined into one or divided.

1 is a schematic diagram of an acoustic signal transmission system including an acoustic compression encoding apparatus and decoding apparatus for multi-channel acoustic signals according to the present invention. 1 is a schematic block diagram of a multi-channel acoustic compression encoding apparatus according to an embodiment of the present invention. It is a schematic block diagram of a multi-channel acoustic category division unit. It is a figure which shows an example of grouping. It is a schematic block diagram of the category 1 acoustic channel compression encoding part. 1 is a schematic block diagram of a multi-channel audio decoding device according to an embodiment of the present invention. It is a schematic block diagram of the category 1 acoustic channel acoustic decoding unit. It is an example of speaker (channel) arrangement at the time of program production in a three-dimensional sound system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Multi-channel acoustic compression encoding apparatus 110 Multi-channel acoustic category division part 112 Multi-channel grouping part 114 Category division part 120 Acoustic compression encoding part 121 Category 1 acoustic channel compression encoding part 121A Time-frequency conversion part 121B Requantization Encoding / Encoding Unit 121C Masking Threshold Calculation Unit 121D Threshold Calculation Unit 121E Acoustic Channel Coding Information Multiplexing Unit 122 Intra-category 2 Acoustic Channel Compression Encoding Unit 12n Intra-category n In-acoustic Channel Compression Encoding Unit 130 Multi-channel Acoustic Multiple Category Code Stream multiplexing unit 140 Category allocation bit rate distribution unit 200 Multi-channel acoustic decoding device 210 Multi-channel acoustic multi-category encoded stream separation unit 220 Acoustic decoding unit 221 Category 1 internal acoustic channel acoustic decoding unit 221A Category 1 internal acoustic channel encoded stream separation unit 221B decoding unit 221C frequency-time conversion unit 230 multichannel acoustic category integration unit 300 3D multi-channel acoustic production and operation device 400 3D Multi-channel sound reproduction apparatus such as sound M1 to M10 Middle layer channel U1 to U9 Upper layer channel L1 to L3 Lower layer channel LFE1, LFE2 Low-frequency sound speaker

Claims (7)

An acoustic compression encoding apparatus for multi-channel acoustic signals by a plurality of acoustic channels arranged apart from each other,
A grouping unit that groups the plurality of acoustic channels based on predetermined arrangement information of the acoustic channels;
For each group grouped by the grouping unit, a category dividing unit that divides the acoustic channels belonging to the group into categories based on the correlation of the acoustic signals of the acoustic channels;
For each category divided by the category dividing unit, each acoustic signal belonging to the category is encoded, and an acoustic compression-by-category that generates an acoustic compression coded signal for each category in which the encoded acoustic signals are multiplexed is generated. An encoding unit;
A multiplexing unit that multiplexes and transmits the category-specific acoustic compression encoded signals generated by the category-specific acoustic compression encoding unit;
An acoustic compression encoding apparatus for multi-channel acoustic signals. The category dividing unit divides the acoustic signals of the acoustic channels belonging to the group into categories for each unit frame for encoding the acoustic signals.
The acoustic compression coding apparatus for multi-channel acoustic signals according to claim 1. A bit rate distribution unit that distributes a predetermined bit rate to each category divided by the category division unit;
The multi-channel acoustic signal acoustic compression coding apparatus according to claim 1 or 2, The bit rate distribution unit distributes a predetermined bit rate according to the number of acoustic channels belonging to each category.
The acoustic compression encoding apparatus for multi-channel acoustic signals according to claim 3, wherein An acoustic decoding device that receives and decodes multi-channel acoustic signals from a plurality of acoustic channels that are spaced apart from each other,
The multi-channel acoustic signal is a multiplexed signal in which acoustic compression-coded signals for each category encoded for each category based on the correlation of the acoustic signal are multiplexed.
A separation unit that separates the multiplexed signal into acoustic compression-coded signals for each category, and extracts category division information related to the separated categories;
A category-specific acoustic decoding unit that decodes the category-specific acoustic compression encoded signals separated by the separation unit for each category and outputs each acoustic signal belonging to the category;
An acoustic signal output by the category sound decoding unit based on the category division information separated by the separating unit integrates, and category integrated unit for output power,
An audio decoding apparatus for multi-channel audio signals, comprising: The acoustic signal is divided into categories for each unit frame for encoding the acoustic signal.
The multi-channel acoustic signal acoustic decoding device according to claim 5, wherein: A predetermined bit rate is allocated to each category,
The multi-channel acoustic signal acoustic decoding device according to claim 5 or 6, wherein

Images